1 Introduction: War-Making, Peacebuilding, and the United Nations

The collapse of state institutions in Somalia, a coup in Haiti, and civil wars in Bosnia, Cambodia, El Salvador, Guatemala, and other countries have marked the distinctive contours of civil strife in the past twenty years. The international community’s responses to these emergencies have been, despite sometimes major efforts, mixed at best: occasional successes in restoring a legitimate and effective government are matched by striking failures to do so.

At the end of the Cold War, the member states of the United Nations (UN) expanded its agenda, defining a near revolution in the relation be­tween what is in the legitimate realm of state sovereignty and what is subject to legitimate international intervention. From 1990 through 1993, the UN Security Council adopted a strikingly intrusive interpreta­tion of UN Charter Chapter VII, the enforcement provisions concerning international peace and security. Member states thus endorsed a radical expansion in the scope of collective intervention just as a series of ethnic and civil wars erupted across the globe. Unfulfilled commitments, on the one hand, and escalating use of force, on the other, soon provoked a se­vere crisis in “peace enforcement.” In Bosnia and Somalia “peace en­forcement” amounted to “war-making” as the United Nations threat­ened to impose by force outcomes—ranging from disarmament, to safe havens, “no fly zones,” and new state borders—on armed factions that recognized no political authority superior to their own.1 Elsewhere, as in Rwanda, the UN record was a failure even to attempt to exercise en­forcement as peace agreements fell apart. As a consequence, more than 700,000 Tutsis and moderate Hutus fell at the hands of genocidal ex­tremists that had seized the government. The current balance sheet on

1 We realize that, and will explain below how, the UN regarded these activities as “peacekeeping” or “peace enforcement,” not war-making. The parties, however, can have reason to see them differently. Imagine, for example, how the U.S. federal government would have viewed a decision of the European Concert in 1864 to establish Washington, Baltimore, Atlanta, Mobile, and New Orleans as “safe havens” and to ban all interference with American commerce in American territorial waters—a “no sail zone”—by either the federal government or the Confederacy. None of this questions whether the UNPROFOR operation in the former Yugoslavia was justified.

2 C H A P T E R 1

UN “war-making” thus suggests that while the UN has served an effec­tive role in legitimizing enforcement coalitions for interstate, armed col­lective security (as in Korea and against Iraq in Gulf War I), the United Nations has proven to be a very ineffective peace enforcer, or war-maker, in the many intrastate, civil conflicts that emerged in the post–Cold War world.

But that is only half the story. At the same time, evidence from the peace operations in Namibia, El Salvador, Cambodia, Mozambique, East­ern Slavonia (Croatia), and East Timor suggests a seemingly contradic­tory (but actually complementary) conclusion. Here the UN succeeded in fostering peace through consent, building on an enhancement of Chapter VI–based peace-making negotiations and a creative, multidimensional implementation of the transitional authority that the peace agreements provided.

Clearly, consent does not guarantee success. The wars in Angola re­futed each of the many agreements that supposedly settled them, and the Rwanda genocide belied the peace agreement signed at Arusha. Weak implementation undermines even the best of agreements. None, more­over, of the successfully implemented operations lacked challenges. In Cambodia the United Nations undertook a multidimensional peace operation—the United Nations Transitional Authority in Cambodia— but the peace it left behind in 1993 was partial as the Khmer Rouge re­sumed sporadic armed resistance. Cambodia also suffered a coup in 1997 and then struggled ahead with an elected government that has been accused of numerous election irregularities. In El Salvador, Guatemala, Namibia, Eastern Slavonia (Croatia), and Mozambique peace is firmer. But even there the long run prospects of social integration remain prob­lematic. In Bosnia, the international community struggles to unite what emerged from the Dayton Peace process as a de facto partition. Current stability is a direct function of the coercive glue of NATO (Stabilization Force) peacekeeping. The international community intervened and as­sumed temporary sovereignty in Kosovo and East Timor. East Timor is now an independent state; the task of assisting the development of a vi­able polity in Kosovo has barely begun.

Despite overcoming many challenges and achieving many successes, the UN’s future as peace-maker has been under challenge in the U.S. Congress and elsewhere from those who fail to understand how success­ful the UN has been and can continue to be in a “peacebuilding” role.2

2 The U.S. Congress, National Security Revitalization Act (H.R. 7) included provisions for charging the UN for a wide range of indirect as well as direct costs of U.S. participation in peacekeeping. If it had been adopted in this form, the legislation (in the eyes of many ex­pert witnesses) would have bankrupted UN peacekeeping as the United States and other states proceeded to charge the UN for what have been extensive voluntary commitments in support of UN peacekeeping efforts. See the testimony of Secretary of State Warren

3 I N T R O D U C T I O N

Obviously, multilateral peacebuilding cannot replace national foreign policy, even in policies directed toward states in crisis. Not only does multilateral peace enforcement regularly fail, but multilateral peace-building, because of its impartial character, will not be the choice that states that seek unilateral advantages will choose. It is not the favored means to impose neo-imperial clients, acquire military bases, or garner economic concessions. Successful multilateral peacebuilding builds func­tioning states that can defend their own interests. But where states seek a sustainable peace to end a festering civil war, multilateral peacebuilding, when well designed and well managed, can produce that peace from which neighbors and the wider international community will benefit, and do so while sharing costs on a fair basis. Clearly, we should avoid “throwing the baby out with the bathwater.”

Stopping civil wars has never been more important. Since the end of the Cold War period, almost all new armed conflicts have occurred within the territories of sovereign states.3 Repeat civil wars in Rwanda and Angola, products of failed peace agreements, alone resulted in several million casualties in the 1990s. Internal (civil or intrastate) war has re­placed interstate war as the paramount concern of organizations charged with maintaining international peace and security. Civil wars have nega­tive security and economic externalities and can destabilize entire re­gions. Beyond the deaths and displacements that are caused directly by the war, civil wars also cause a deterioration of health levels for the en­tire region long after the fighting ends.4 Civil wars have regional conta­gion or diffusion effects,5 and they reduce rates of economic growth in both the directly affected countries and their neighbors.6 Civil wars typi­cally do not occur between standing armies, but rather between a gov­ernment army, or militia, and one or more rebel organizations. Violence

Christopher (Thursday, January 26, 1995) and C. William Maynes (January 19, 1995) be­fore the House International Relations Committee.

3 There have been few interstate wars, including the war between Ethiopia and Eritrea— which is itself the continuation of an earlier internal war—the Gulf War in 1991, the U.S. invasion of Afghanistan and Iraq in 2001 and 2004, respectively, and the war in the Demo­cratic Republic of the Congo, which involved many neighboring states and had a large civil war component.

4 Hazem Ghobarah, Paul Huth, and Bruce Russett, “Civil Wars Kill and Maim People, Long after the Fighting Stops,” American Political Science Review 97 no. 2 (2003): 189–202.

5 Michael E. Brown, “The Causes and Regional Dimensions of Internal Conflict,” in In­ternational Dimensions of Internal Conflict (Cambridge: MIT Press, 1996), 571–602; D. A. Lake and D. Rothchild, eds., The International Spread of Ethnic Conflict: Fear, Diffu­sion, and Escalation (Princeton: Princeton University Press, 1998); Nicholas Sambanis, “Do Ethnic and Non-ethnic Civil Wars Have the Same Causes? A Theoretical and Empiri­cal Inquiry (part 1),” Journal of Conflict Resolution 45, no. 3 (2001): 259–82.

6 James C. Murdoch and Todd Sandler, “Economic Growth, Civil Wars, and Spatial Spillovers,” Journal of Conflict Resolution 46 (February 2002): 91–110.

4 C H A P T E R 1

is usually targeted at civilians, and the objectives of civil wars range from secession to control of the state or resource predation. Civilian deaths as a percentage of all war-related deaths increased to 90 percent in 1990 from approximately 50 percent in the eighteenth century. Inter­nal wars have created approximately 13 million refugees and 38 million internally displaced persons.7

This book will discuss theories of the origins of and solutions to civil wars, the principles behind and the practices of the United Nations as an institution, and the debate over doctrines and strategies of inter­vention. But its key purpose is to explain how the international commu­nity, and the UN in particular, can assist the reconstruction of peace in civil war–torn lands. We address the policy problem, but we assess it in ways that draw on and apply relevant theories and methods in political science.

We focus on the international role in peacebuilding, even though it is only part of what makes for success or failure. We will argue that “sus­tainable peace” is the measure of successful peacebuilding. Our central claim is that successful and unsuccessful efforts to resolve civil wars are influenced by three key factors that characterize the environment of the postwar civil peace:

1. the degree of hostility of the factions (measured in terms of human cost—deaths and displacements—the type of war, and the number of fac­tions);

2. the extent of local capacities remaining after the war (measured, for ex­ample, in per capita GDP or energy consumption); and

3. the amount of international assistance (measured in terms of economic assistance or the type of mandate given to a UN peace operation and the num­ber of troops committed to the peace effort).

Together, these three constitute the interdependent logic of a “peace­building triangle”: the deeper the hostility, the more the destruction of local capacities, the more one needs international assistance to succeed in establishing a stable peace.8 We find support for this hypothesis both in our case studies and in our statistical analysis of all civil wars since 1945. Controlling for levels of hostility and local capacities, we find that the international capacities—UN missions with a mandate and resources to build peace—increase the chance for peace after civil war.

7 United Nations High Commissioner for Refugees, “State of the World’s Refugees, 1997–98,” cited in Michael W. Doyle and Anne Bayefski, “Sustainable Refugee Return: A Report of a Workshop at Princeton University,” Unpublished paper, Princeton University (February 1998).

8 Michael Doyle and Nicholas Sambanis, “International Peacebuilding: A Theoretical and Quantitative Analysis,” American Political Science Review 94, no. 4 (2000): 779–801.

5 I N T R O D U C T I O N

We find that peace operations must be designed to fit the case, with the kind and degree of international authority to shape the transition from war to peace. The valuable monitoring that can be sufficient to re­inforce trust and serve as a midwife to peace in one case is the idle ob­server that merely witnesses the collapse of a peace among hostile fac­tions in a second case that would have required robust transitional executive authority for success.

We further find that peace operations supplemented by extensive pro­grams to rebuild economies have a particularly prominent role in promot­ing long-run peace. Peacebuilding requires the provision of temporary security, the building of new institutions capable of resolving future con­flicts peaceably, and an economy capable of offering civilian employ­ment to former soldiers and material progress to future citizens.

Peacebuilding, however, does not require that the United States, or an­other great power, take the lead. When residual violence is plentiful, such leadership may be necessary. In less violent circumstances, however, multilateralism works well, delivering the legitimacy, staying power, ex­perienced UN peacekeepers, and multiple sources of modest national commitment that it promises.

Lastly, controversially, we find that peacebuilding trumps military vic­tories. Most civil wars since World War Two have been settled by military victory, and these victories can deliver a stable peace by eliminating the organized military opposition that truces leave in place to stir up future trouble. But a comprehensive peace agreement implemented through a peace operation has an even better success rate.

Our policy message is simple: while the UN is very poor at “war,” im­posing a settlement by force, it can be very good at “peace,” mediating and implementing a comprehensively negotiated peace. This will not shock the insiders. What is new in this book is demonstrating this asser­tion carefully and explaining why and how this is the case. In exploring “why” we argue that the UN, as a multilateral organization, cannot manage force as rationally as is necessary but it is well suited to mediate, mobilize, and manage legitimate international assistance. These institu­tional capacities reflect wider views on the illegitimacy of colonialism and the growing acceptability of peacekeeping and peacebuilding.9

9 That is not to say that multilateral management guarantees either good or selfless management. Peacekeeping operations can be, and have been, exploited for private or na­tional gain, as the “Oil for Food” investigation has alleged (July 2005). Ordinary peace­keeping operations generate opportunities for profit, as Michael Bhatia has illustrated in “Postconflict Profit: The Political Economy of Intervention,” Global Governance 11 (2005): 205–24. Multilateralism, nonetheless, has value in establishing internationally agreed common ground, mitigating the exploitation of simple national advantage and en­hancing transparency. In 2003–4, for example, the importance attached by the United

6 C H A P T E R 1

In explaining “how” we identify the sources of failures in UN war-making and explore the four innovations (enhanced forms of peacemak­ing, peacekeeping, peacebuilding, and “discrete enforcement”) that led to success. And we describe how the authority embedded in peacebuild­ing operations must be tailored to the circumstances they face.

These conclusions are important, partly because the use of UN au­thorized peace operations greatly increased in the 1990s, reflecting a new wave of interventionism and redefining a new generation of strate­gies in peacekeeping designed to fulfill the ambitious expectations un­leashed by the new willingness to intervene. The connections between interventionism, new strategies, and successful peacebuilding were inti­mate and serious: no matter how well intentioned an intervention is, un­less the intervenor can also claim that the intervention is likely to produce a sustainable improvement—both peace and human rights—the interven­tion is unlikely to be either ethically justifiable or politically viable.

The New Interventionism

As Secretary-General Kofi Annan memorably described the new UN role in 1998: “Our job is to intervene: to prevent conflict where we can, to put a stop to it when it has broken out, or—when neither of those things is possible—at least to contain it and prevent it from spreading.”10 He was reflecting the activism of the Security Council, which between 1987 and 1994 had quadrupled the number of resolutions it issued, tripled the peacekeeping operations it authorized, and multiplied by seven the num­ber of economic sanctions it imposed per year. Military forces deployed in peacekeeping operations increased from fewer than 10,000 to more than 70,000. The annual peacekeeping budget skyrocketed correspond­ingly from $230 million to $3.6 billion in the same period, thus reaching to about three times the UN’s regular operating budget of $1.2 billion.11

States and its coalition of allies in Iraq in both securing international legitimacy and mobi­lizing international assistance in peacebuilding there are one measure of this. One frequent critic of the UN, William Safire, even went so far as to acknowledge that the contrast be­tween the successful elections in Afghanistan in October 2004 and the escalating crisis in Iraq could partly be attributed to the multilateral legitimacy the former enjoyed, in William Safire “The Best Political News of 2004,” New York Times, October 26, 2004.

10 Kofi Annan, “Reflections on Intervention,” Ditchley Park, UK, June 26, 1998, in Kofi Annan, The Question of Intervention (New York: United Nations, 1999) p. 4.

11 Boutros Boutros-Ghali, Supplement to “An Agenda for Peace”: Position Paper of the Secretary-General on the Occasion of the Fiftieth Anniversary of the United Nations, A/50/60; S/1995/1, January 3, 1995, p. 4.

7 I N T R O D U C T I O N

The activities of the Security Council in preventive diplomacy and sanc­tions, the Secretariat’s role in election monitoring, and above all, the massive growth in peacekeeping and peace enforcement all testified to the newly appreciated role the international community wanted the UN—or somebody—to play.

The international legal prohibitions against intervention were more relevant than ever given the demands for national dignity made by the newly independent states of both the Third World and the former Sec­ond World. But the rules as to what constitutes intervention and what constitutes international protection of basic human rights shifted as well. Sovereignty was redefined to incorporate a global interest in human rights protection. The traditional borders between sovereign consent and intervention were blurred. Peacekeeping and peace enforcement almost merged into “robust peacekeeping,” which signaled a willingness to use force if needed whether in consent-based peacekeeping or imposed peace enforcement. A newly functioning United Nations, moreover, was seen to be a legitimate agent to decide when sovereignty was and was not vio­lated.

The revival of the UN Security Council led to a reaffirmation after years of Cold War neglect of the UN Charter’s Article 2, clause 7 affirm­ing nonintervention, except as mandated by the Security Council under Chapter VII. The UN then claimed a “cleaner hands” monopoly on le­gitimate intervention. Although the letter of the Charter prohibited UN authorizations of force other than as a response to threats or breaches of “international” peace, the Genocide Convention and the record of con­demnation of colonialism and apartheid opened an informally legitimate basis for involvement in domestic conflict. The Security Council’s prac­tice thus broadened the traditional reasons for intervention, including aspects of domestic political oppression short of massacre and human suffering associated with economic misfeasance—the so-called failed states and the droit d’ingerence.12 Building on new interpretations ad­vanced during the Cold War that made, for example, apartheid a matter for international sanction, the United Nations addressed the starvation of the Somali people when it became clear that its government was inca­pable of doing so. (In this case, however, the traditional criteria of “in­ternational” threats were also invoked—including Somali refugees spread­

12 See Lori F. Damrosch, ed., Enforcing Restraint: Collective Intervention in Internal Con­flicts (New York: Council on Foreign Relations Press, 1993): G. Helman and S. Ratner, “Saving Failed States,” Foreign Policy 89 (Winter 1992–93): 3–20: but for a more skeptical reading, Edward Mortimer, “Under What Circumstances Should the UN Intervene Militarily in a ‘Domestic’ Crisis?” in Olara Otunnu and Michael Doyle, eds., Peacemaking and Peace­keeping for the New Century (Lanham, MD.: Rowman and Littlefield, 1998), pp 111–44.

8 C H A P T E R 1

ing across international borders—in order to justify forcible intervention under Chapter VII.) The Security Council also demanded international humanitarian access to vulnerable populations, insisting, for example, that humanitarian assistance be allowed to reach the people affected in Yugoslavia and in Iraq.13

Regions differed on the meaning of operational sovereignty. The Asso­ciation of Southeast Asian nations (ASEAN) remained a bastion of strict sovereignty, and nonintervention is the norm. Although Cambodia and Burma’s acceptance into ASEAN were delayed by their human rights record and instability, they were both eventually accepted. The Organiz­ation of African Unity (OAU), on the other hand has defined standards of (1990) “Good Governance” that included democracy and declared (July 3, 1993) that internal disputes are matters of regional concern. And, more strikingly, the Organization of American States (in Res. 1080 and in the “Santiago Commitment of 1991”) has declared coups against democracy illegitimate and has adopted economic sanctions against coups in Haiti and Peru. The European Union makes democracy an element in the criteria it demands for consideration in membership.

It was also important that the “international community” had a newly legitimate means of expressing its collective will on an internationally im­partial basis. The Security Council lays claim to being the equivalent of a “global parliament” or “global jury”14 representing not merely the indi­vidual states of which it is composed but also a collective will and voice

13 In an important recent report, the International Commission on Intervention and State Sovereignty has affirmed and called upon the Security Council to recognize “a re­sponsibility to protect.” International Commission on Intervention and State Sovereignty, The Responsibility to Protect [Ottawa: International Development Research Centre, 2001]. States of course have the first responsibility to protect the basic rights and welfare of their citizens, but if they should fail to do so through lack of will or capacity, the re­sponsibility should devolve, the commission argues, onto the international community, with the Security Council as its agent. Widely discussed, though not formally endorsed at the United Nations in 2002, the report sets a new benchmark against which future inter­ventions and noninterventions will be judged. The report, however, deals less well with a separate problem: What should happen when the Security Council is deadlocked? Michael Walzer, in “The Politics of Rescue,” Dissent (Winter 1995), has persuasively argued that the Security Council should not have the last word, when it comes up with the wrong an­swer. Tom Farer has explored the circumstances under which the responsibility to inter­vene devolves from a deadlocked Council to regional organizations and national govern­ments. See Tom J. Farer, “Humanitarian Intervention before and after 9/11: Legality and Legitimacy,” in J. L. Holzgrefe and Robert Keohane, eds., Humanitarian Intervention: Ethical, Legal, and Political Dilemmas (Cambridge: Cambridge University Press, 2003).

14 Farer 2003; and Thomas Franck, “Interpretation and Change in the Law of Humani­tarian Intervention,” in J. L. Holzgrefe and Robert Keohane, eds., Humanitarian Interven­tion: Ethical, Legal, and Political Dilemmas (Cambridge: Cambridge University Press, 2003), p. 227.

9 I N T R O D U C T I O N

of the “international community.” The Security Council includes five per­manent members (United States, Russia, France, the United Kingdom, and China) and ten nonpermanent, elected members, always including members from Asia, Africa, and Latin America. Its authorization for an intervention requires the affirmative vote of nine states, including no neg­ative votes from the five permanent members (the P5) and four positive votes from the ten elected members. Such a vote would have to incorpo­rate representatives of a variety of cultures, races, and religions. It would always include representatives of large and small countries, capitalist and socialist economies, and democratic and nondemocratic polities. If the mandated operation is UN directed and if troops and funding are re­quired, many other troop contributing states will be needed, and they can say no in practice. The combination makes for a genuinely international impartial intervention, and hence “cleaner hands.”

Those developments coincided with a temporary conjunction of power and will. Following the collapse of the USSR, the United States experi­enced a “unipolar moment” when its power eclipsed that of all other states. At the same time the international community, including the United States, adopted a strategy of “assertive multilateralism,” which lasted from the Gulf War in January 1991 until the October 3, 1993, disaster in Mogadishu, Somalia. The Five Permanent Members of the Security Council, led by the United States, provided a degree of commitment and resourceful leadership that the UN had rarely seen before. Eschewing the national role of “Globocop” in order to address a pressing domestic agenda, the Clinton administration encouraged then UN Secretary-General Boutros Boutros-Ghali to take an ever more assertive role in in­ternational crises. The small dissenting minority in the Security Council— which included China on some occasions and Russia on others—was not prepared to resist the United States on issues that did not affect their paramount national interests. The successful reversal of Saddam Hus­sein’s aggression in the Gulf and the December 1992 U.S.-led rescue of segments of the Somali population from starvation heralded what ap­peared to be a remarkable partnership. The Security Council decreed, the United States led, and—conveniently, for the while—many other states paid and supported.

But cleaner hands need not mean better hands. The international com­munity still needed to find a way to promote sustainable peace, one that enlisted the support of a substantial majority of the local population and embodied basic principles of human rights. This proved to be a challenge in the 1990s as massive UN interventions, warlike enforcement opera­tions, provoked some of the same sorts of resistance as did colonial in­terventions in places such as Somalia and Bosnia. But in other cases the UN found a way to cultivate consent and then build a peace with gen­

10 C H A P T E R 1

uine indigenous roots. These peacebuilding operations rested on impor­tant innovations in peacemaking, peacekeeping, and institutional recon­struction, as well as discrete residual enforcement, all of which evolved to address particular aspects of these challenges.

Generations of UN Peace Operations

In the early 1990s, with the end of the Cold War, the UN’s agenda for peace and security thus rapidly expanded. At the request of the UN Se­curity Council Summit of January 1992, then Secretary-General Boutros Boutros-Ghali prepared the conceptual foundations of an ambitious UN role in peace and security in his seminal report, An Agenda for Peace (1992).15 In addition to preventive diplomacy designed to head off con­flicts before they became violent, the Secretary General outlined the four interconnected roles that he hoped the UN would play in the fast chang­ing context of post–Cold War international politics.

• preventive diplomacy, undertaken in order “to prevent disputes from aris­ing between parties, to prevent existing disputes from escalating into con­flicts and to limit the spread of the latter when they occur.” Involving confidence-building measures, fact-finding, early warning and possibly “preventive deployment” of UN authorized forces, preventive diplomacy seeks to reduce the danger of violence and increase the prospects of peace­ful settlement.

• peace enforcement, authorized to act with or without the consent of the parties in order to ensure compliance with a cease-fire mandated by the Se­curity Council acting under the authority of Chapter VII of the UN Char­ter, these military forces are composed of heavily armed national forces operating under the direction of the Secretary-General.

• peacemaking, designed “to bring hostile parties to agreement” through peaceful means such as those found in Chapter VI of the UN Charter. Drawing upon judicial settlement, mediation, and other forms of negotia­tion, UN peacemaking initiatives would seek to persuade parties to arrive at a peaceful settlement of their differences.

• peacekeeping, established to deploy a “United Nations presence in the field, hitherto with the consent of all the parties concerned,” as a confidence-building measure to monitor a truce between the parties while diplomats strive to negotiate a comprehensive peace or officials to implement an agreed peace.

15 Boutros Boutros-Ghali, An Agenda for Peace: Report of the Secretary-General, A/47/277-S/24111 (June 17, 1992), http://www.un.org/Docs/SG/agpeace.html. Quotes that follow are from paragraphs 20–21 and 55–99.

11 I N T R O D U C T I O N

• postconflict reconstruction,16 organized to foster economic and social cooperation with the purpose of building confidence among previously war­ring parties, developing the social, political, and economic infrastructure to prevent future violence, and laying the foundations for a durable peace.

The Secretary-General’s Agenda for Peace is the culmination of an evolution of UN doctrine and an adjustment of the instruments used to maintain the peace since the organization was formed in 1945. It com­bines in a radical way instruments of warlike enforcement and peacelike negotiation that were once kept separate and that evolved separately. A unique vocabulary separates distinct strategies that fit within the generic UN doctrine of building peace. These strategies, evolving over time, have encompassed three generational paradigms of peacebuilding.17 They in­clude not only the early activities identified in UN Charter Chapter VI (or so-called 6 and 1/2)18 first generation peacekeeping, which calls for the interposition of a force after a truce has been reached, but also a far more ambitious group of second generation operations that rely on the consent of parties and an even more ambitious group of third generation operations that operate with Chapter VII mandates and without a com­prehensive agreement reflecting the parties’ acquiescence. In today’s cir­cumstances, these operations involve less interstate conflict and more factions in domestic civil wars, not all of whom are clearly identifiable— and few of whom are stable negotiating parties. Current peace opera­tions thus intrude into aspects of domestic sovereignty once thought to be beyond the purview of UN activity.

Indeed, the post–World War Two UN Charter can be seen as having been designed for interstate wars (e.g., Article 39’s threats to “interna­tional” peace); appropriately so, since, from 1900 to 1941, 80 percent of all wars were interstate among state armies. But from 1945 to 1976, 85 percent of all wars were on the territory of one state and internally ori­ented—of course with proxies.19

16 The Secretary-General and the UN often refers to this as “post-conflict peacebuild­ing.” To avoid confusion with the wider meaning of peacebuilding we employ, we will call it postconflict reconstruction.

17 It is worth recalling that the time line of evolution has by no means been chronologi­cally straightforward. One of the most extensive “third generation” operations undertaken by the UN was ONUC in the then-Congo, from 1960 to 1964, which preceded the spate of “second generation” operations that began with UNTAG in Namibia in 1989.

18 The “6 and 1/2” refers to the fact that peacekeeping per se is nowhere described in the Charter and thus falls between Chapter VI, peacemaking (good offices, etc.), and Chapter VII, peace enforcement.

19 Ernst B. Haas, The United Nations and Collective Management of International Con­flict (New York: UNITAR, 1986); and Henry Wiseman, “The United Nations and Interna­tional Peace,” in UNITAR, The United Nations and the Management of International Peace and Security (Lancaster: Martinus Nijhof, 1987), 219.

12 C H A P T E R 1

Traditional peace operations, or first generation peacekeeping, were designed to respond to interstate crises by stationing unarmed or lightly armed UN forces between hostile parties to monitor a truce, troop with­drawal, or buffer zone while political negotiations went forward.20 As F. T. Liu, an eminent peacekeeping official of the UN has noted: monitor­ing, consent, neutrality, nonuse of force, and unarmed peacekeeping—the principles and practices of first generation peacekeeping—constituted a stable and interdependent combination. These key principles were artic­ulated by Secretary-General Dag Hammarskjold and former Canadian prime minister Lester Pearson in conjunction with the creation of the first peacekeeping operation, the UN Emergency Force (UNEF) in the Sinai, which was sent to separate Israel and Egypt following the Franco-British-Israeli intervention in Suez in 1956.21 The principle of neutrality referred to the national origin of UN troops and precluded the use of troops from the permanent five members of the Council in order to quiet fears of superpower intervention. Impartiality implied that the UN would not take sides in the dispute and was a precondition for achieving the consent of all the parties. Enjoying the consent of all fac­tions in turn made it easier for monitors of peacekeepers not to have to use force except in self-defense.22 Lastly, the Secretary-General exercised control of the force and the Security Council authorized it (or rarely, the General Assembly under the auspices of the “Uniting for Peace Res­olution”).23

20 The first peacekeeping operation was the United Nations’ Emergency Force (UNEF) in Egypt, deployed in October 1956 to maintain a truce between the Egyptian army and Is­rael, England, and France during the Suez crisis. UNEF’s experience helped define the four principles of traditional peacekeeping: consent, impartiality, neutrality, and use of force only in self-defense. The UN Treaty Supervision Organization (UNTSO) was deployed in 1948 in Palestine, but it was a limited observer mission.

21 United Nations, The Blue Helmets, 2d ed. (New York: United Nations, 1990), 5–7; and Brian Urquhart, A Life in Peace and War (London: Weidenfeld and Nicolson, 1987), p. 133.

22 Traditional peacekeeping is a shorthand term that describes many but by no means all Cold War peacekeeping missions (the most notable exception being the Congo operation and possibly also the Cyprus operation, as we discuss later in the book). For cogent analy­ses of different types of peacekeeping, see Marrack Goulding, “The Evolution of United Nations Peacekeeping,” International Affairs 69, no. 3 (1993): 451–64; F. T. Liu, United Nations Peacekeeping and the Nonuse of Force, International Peace Academy Occasional Paper Series (Boulder, CO: Lynne Rienner, 1992); Thomas G. Weiss, ed., Collective Secu­rity in a Changing World (Boulder CO: Lynne Rienner, 1993).

23 A controversial resolution introduced in the context of the Korean War designed to circumvent the deadlock in the Security Council that resulted from the return of the USSR to the Council, following the boycott that allowed the Council in the USSR’s absence to authorize the U.S.-led force in Korea in June 1950. It was applied to authorize the Sinai peace force in 1956.

13 I N T R O D U C T I O N

Impartiality and neutrality are frequently used interchangeably. Schol­ars and practitioners often speak of peacekeepers as “neutral,” “disinter­ested,” “impartial,” or “unbiased,” and they tend to mistake the need for impartiality with a policy of “strict neutrality” and a disposition of pas­sivity. In this book, we define neutrality as a synonym for noninterfer­ence with respect to peacekeeping outcomes and impartiality as equal enforcement of unbiased rules. Good cops act impartially but not neu­trally when they stop one individual from victimizing another. We argue that it is as important for peacekeepers to be impartial concerning, for example, which party in a freely conducted democratic election wins the election as it is for them to be nonneutral (i.e., not passive) with respect to violations of the peace and obstructions to their ability to implement their mandate. This is closely related to the interpretation of the fourth principle of peacekeeping—the nonuse of force. Peacekeeping uses sol­diers not to win wars, but rather to preserve the peace. But peacekeepers must also protect their right to discharge their functions, in accordance with the spirit of the parties’ consent as extended at the outset of the op­eration. Raising the costs of noncooperation for the parties must, on oc­casion, allow the use of force in defense of the mandate. The limited use of force to protect a mandate authorized by a peace treaty or to enforce an agreed-upon cease-fire (as happened in Cyprus in 1974 or Namibia in 1989), does not equate peacekeeping with peace enforcement (which at­tempts to impose an overall settlement), but it does generate concerns with mission creep if the need to use force is extensive.

During the Cold War, the UN record indicated much success in in­terstate conflicts (while little in intrastate) and much in material and ter­ritorial settlement (while little in value or identity conflicts).24 The success of traditional peacekeeping was also dependent on successful peacemak­ing: a strategy designed “to bring hostile parties to agreement” through peaceful means such as those found in Chapter VI of the UN Charter. Drawing upon judicial settlement, mediation, and other forms of negoti­ation, UN peacemaking initiatives would seek to persuade parties to ar­rive at a peaceful settlement of their differences. Traditional peacekeep­ing operations referred to the deployment of a United Nations presence in the field, with the consent of all the parties concerned, as a confidence building measure to monitor a truce while diplomats negotiated a com­prehensive peace. Peacekeeping was therefore designed as an interim arrangement where there was no formal determination of aggression, and was frequently used to monitor a truce, establish and police a buffer

24 Hugh Miall, The Peacemakers: Peaceful Settlement of Disputes since 1945 (New York: St. Martin’s, 1992), p. 185, 112–13; Paul Diehl, International Peacekeeping (Balti­more: Johns Hopkins University Press, 1993), p. 171.

14 C H A P T E R 1

zone, and assist the negotiation of a peace. Monitoring or observer mis­sions had several of the same objectives as traditional peacekeeping op­erations, though they were typically less well armed (or unarmed) and focused on monitoring and reporting to the Security Council and the Secretary-General.

Both monitoring operations and traditional peacekeeping provided transparency—an impartial assurance that the other party was not vio­lating the truce—and were supposed to raise the costs of defecting from an agreement by the threat of exposure and the potential (albeit un­likely) resistance of the peacekeeping force. The international legitimacy of UN mandates increased the parties’ benefits of cooperation with the peacekeepers. The price of first generation peacekeeping, as in the long Cyprus operation, was sometimes paid in conflicts delayed rather than resolved. Today these monitoring activities continue to play an impor­tant role on the Golan Heights between Israel and Syria and, until re­cently, on the border between Kuwait and Iraq.

Monitoring and traditional peacekeeping operations were strictly bound by the principle of consent. Consent derives from the parties’ “perceptions of the peacekeepers’ impartiality and moral authority.”25 It reduces the risk to the peacekeepers and preserves the sovereignty of the host state. Eroding consent can significantly diminish the peacekeepers’ ability to discharge their mandate, so the peacekeepers have an incentive to enhance the parties’ consent. Since eroding consent could turn PKOs into multibillion-dollar “obsolescing investments” that are easy hostages to insincere parties, it follows that the UN should develop strategies to enhance consent.26 This flexibility is more easily provided in second gen­eration, multidimensional operations that involve the implementation of complex, multidimensional peace agreements designed to build the foun­dations of a self-sustaining peace and have been utilized primarily in post–civil war situations. In addition to the traditional military func­tions, the peacekeepers are often engaged in various police and civilian tasks, the goal of which is a long-term settlement of the underlying con­flict. These operations are based on the consent of the parties, but the nature of and purposes for which consent is granted are qualitatively dif­ferent from traditional peacekeeping.

In addition to monitoring and traditional peacekeeping, the key strat­

25 William J. Durch, The Evolution of UN Peacekeeping: Case Studies and Comparative Analyses (New York: St. Martin’s Press, 1993), p. 12.

26 Michael W. Doyle, UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate (Boul­der, CO: Lynne Rienner Publishers, 1995), p. 85; Steven R. Ratner, The New UN Peace­keeping: Building Peace in Lands of Conflict after the Cold War (New York: St. Martin’s Press, 1995), p. 39.

15 I N T R O D U C T I O N

egy was to foster economic and social cooperation with the purpose of building confidence among previously warring parties, developing the social, political, and economic infrastructure to prevent future violence, and laying the foundations for a durable peace. Multidimensional peace­keeping is aimed at capacities expansion (e.g., economic reconstruction) and institutional transformation (e.g., reform of the police, army, and ju­dicial system, elections, civil society rebuilding). In these operations, the UN is typically involved in implementing peace agreements that go to the roots of the conflict, helping to build long-term foundations for sta­ble, legitimate government. As Secretary-General Boutros-Ghali observed in An Agenda for Peace, “peace-making and peace-keeping operations, to be truly successful, must come to include comprehensive efforts to identify and support structures which will tend to consolidate peace. . . . [T]hese may include disarming the previously warring parties and the restoration of order, the custody and possible destruction of weapons, repatriating refugees, advisory and training support for security person­nel, monitoring elections, advancing efforts to protect human rights, re­forming or strengthening governmental institutions, and promoting for­mal and informal processes of political participation.”27

The UN has a commendable record of success, ranging from mixed to transformative, in “second generation,” multidimensional peace opera­tions as diverse as those in Namibia, El Salvador, Cambodia, Mozam­bique, and Eastern Slavonia (Croatia).28 The UN’s role in helping settle those conflicts has been fourfold. It served as a peacemaker facilitating a peace treaty among the parties; as a peacekeeper monitoring the canton­ment and demobilization of military forces, resettling refugees, and su­pervising transitional civilian authorities; as a peacebuilder monitoring and in some cases organizing the implementation of human rights, na­tional democratic elections, and economic rehabilitation; and in a very limited way as peace enforcer when the agreements came unstuck.

In Secretary General Boutros-Ghali’s lexicon, “peace-enforcing”— effectively war-making—missions are third generation operations, which extend from low-level military operations to protect the delivery of hu­manitarian assistance to the enforcement of cease-fires and, when neces­sary, authoritative assistance in the rebuilding of so-called failed states. Like Chapter VII UN enforcement action to roll back aggression, as in Korea in 1950 and against Iraq in the Gulf War, the defining characteris­tic of “third generation” operations is the lack of consent by one or

27 Boutros-Ghali 1992, para. 21. 28 Success is of course an ambiguous and contested term. We explain later how we de­

fine and measure it in our statistical analysis and case studies.

16 C H A P T E R 1

more of the parties to some or all of the UN mandate.29 These operations have been of three types. In the first, international forces attempt to im­pose order without significant local consent, in the absence of compre­hensive peace agreement, and must in effect conquer the factions (as was attempted in Somalia). In the second, international forces did not have unanimous consent and have chosen to impose distinct arrangements on parties in the midst of an ongoing war (e.g., no-fly zones or humanitar­ian corridors of relief ). In the third, international forces exercise force to implement the terms of comprehensive peace from which one or more of the parties has chosen to defect.

Enforcement operations draw upon the authority of UN Charter Arti­cle 42, which permits the Security Council to “take such action by air, sea, or land forces as may be necessary to maintain or restore interna­tional peace and security”; Article 25 under which member states “agree to accept and carry out the decisions of the Security Council”; and Arti­cle 43 in which they agree to “make available to the Security Council, on its call, . . . armed forces, assistance and facilities.”

Insightful doctrine for these peace-enforcing operations appeared just as Somalia and Bosnia exposed their practical limitations. Recent studies have thoughtfully mapped out the logic of the strategic terrain between traditional UN peacekeeping and traditional UN enforcement action. Militarily, these operations seek to deter, dissuade, and deny.30 By pre­cluding an outcome based on the use of force by the parties, the UN in­stead uses collective force (if necessary) to persuade the parties to settle the conflict by negotiation. In the former Yugoslavia, for example, the UN following this strategy could have established strong points to deter attacks on key humanitarian corridors. (It actually did, but the Serbs by­passed them.) Or it could threaten air strikes, as was done successfully around Sarajevo in February 1994, to dissuade a continuation of the Serb shelling of the city. Or it could have denied (but did not) the Serb

29 Other recent categories include “preventive deployments” deployed with the intention of deterring a possible attack, as in the Former Yugoslav Republic of Macedonia. There the credibility of the deterring force must ensure that the potential aggressor knows that there will be no easy victory. In the event of an armed challenge, the result will be an inter­national war that involves costs so grave as to outweigh the temptations of conquest. Enforcement action against aggression (Korea or the Gulf ), conversely, is a matter of achieving victory—“the decisive, comprehensive and synchronized application of prepon­derant military force to shock, disrupt, demoralize and defeat opponents”—the tradi­tional zero-sum terrain of military strategy. See John Mackinlay and Jarat Chopra, “Sec­ond-Generation Multinational Operations,” Washington Quarterly 15 (Summer 1992), pp. 113–31.

30 See John Ruggie, “The United Nations Stuck in a Fog between Peacekeeping and Peace Enforcement,” McNair Paper 25 (Washington, DC: National Defense University, 1993), for these distinctions.

17 I N T R O D U C T I O N

forces their attack on Dubrovnik in 1992 by countershelling from the sea or bombing from the air of the batteries in the hills above the city. Forcing a peace depends on achieving a complicated preponderance in which the forces (UN and local) supporting a settlement acceptable to the international community hold both a military predominance and a predominance of popular support, which together permit them to im­pose a peace on the recalcitrant local military forces and their popular supporters.

Countries provide troops to UN peace operations in various ways. Troop-contributing countries negotiate in detail the terms of the partici­pation of their forces either under UN command and thus with the Secretary-General (as in El Salvador or Cambodia); with a regional or­ganization authorized as delegated in Chapter VIII; or with the leader of a multinational “coalition of the willing” authorized under Chapter VII, as was the case of U.S. leadership of Unified Task Force (UNITAF, sand­wiched in between the two UN operations in Somalia). Many operations draw on a combination of authorizations: peace treaties among factions, backed up or supplemented by other measures authorized (such as arms embargoes, no-fly zones) under Chapter VII, as did the various UN Pro­tection Force (UNPROFOR) and NATO Implementation Force (IFOR) operations in the former Yugoslavia.31 And, as named in honor of its sponsors, “Chinese Chapter Seven” (employed to authorize the use of force for UNTAES) has emerged as a new way to signal firm intent to en­force a Chapter Six operation. In essence, however, it reaffirms the “Katanga Rule” of the ONUC operation in the Congo: the traditional principle that force can be used both in self-defense of peacekeeping troops and of the mission (mobility of the force).

The result of these three “generations” operating together in the post–Cold War world was an unprecedented expansion of the UN’s role in the protection of world order and in the promotion of basic human rights in countries, until recently, torn by costly civil wars. Self-determination and sovereignty were enhanced and a modicum of peace, rehabilitation, and self-sustaining self-determination was introduced in Namibia, Cam­bodia, El Salvador, Mozambique, and Eastern Slavonia. Tens—perhaps, even hundreds—of thousands of lives were saved in Somalia and the for­mer Yugoslavia. But in 1993 and 1994, the more ambitious elements of “third generation” peace enforcement encountered many of the problems interventionist and imperial strategies have faced in the past, and discov­ered fresh problems peculiar to the UN’s global character.

31 For a valuable discussion of the international law on the use of force and its bearing on authority for peace operations see Karen Guttieri’s “Symptom of the Moment: A Juridical Gap for US Occupation Forces,” International Insights 13, special issue (Fall 1997): 131–55.

18 C H A P T E R 1

The debacles in Somalia and Bosnia forced a radical rethinking of when and where the UN should get involved. Disingenuously, President Clinton told the General Assembly that it needed to learn when to say no. Many came to believe that the UN was not well suited to mounting effective peace operations—no more suited to make peace than the lob­byists who represented a trade group of hospitals would be to conduct surgery.32 Others thought that such operations should be delegated to re­gional organizations, and NATO preeminently. This last group began calling for a “fourth generation” of delegated peacekeeping.33

The Challenge of Peacebuilding

The United Nations, as we will argue, has been and can continue to be effective at peace operations, provided it takes to heart the true meaning of its successes and failures. This is not a straightforward task.

Measuring successful peace is a complicated substantive and method­ological issue and much debated in the literature. Many use the Corre­lates of War (COW) definition of peace (fewer than 1,000 battle deaths per annum).34 We adopt a similar standard as one measure of peace, which we call “negative” or “sovereign” peace, reflecting that single sov­ereignty, a Hobbesian Leviathan, has been reestablished and exercises a legitimate monopoly of violence. We add to that standard measure a sec­ond definition of peace. The second definition is “positive,” or “partici­patory” peace, which discounts “peaces of the grave” (the former enemy is all dead or in prison) in favor of a peace that includes wider participa­tion.35 We add data from the Polity project to code a minimal degree of political assent and participation.36 We add this second definition for

32 Michael Mandelbaum, “The Reluctance to Intervene,” Foreign Policy 95 (Summer 1994): 3–18.

33 For an account of the various positions and factors, see Ramesh Thakur, “UN Peace Operations and US Unilateralism and Multilateralism,” in David Malone and Yuen F Kong, eds., Unilateralism and US Foreign Policy (Boulder, CO: Lynne Rienner, 2003), pp. 153–79.

34 See, for example, Barbara Walter, Committing to Peace: The Successful Settlement of Civil Wars (Princeton: Princeton University Press: 2002).

35 In our previous work (Doyle and Sambanis 2000) we called “sovereign” peace “le­nient” peacebuilding and “participatory” peace “strict” peacebuilding. We now prefer the more descriptive terms that rest on the distinction drawn by Kenneth Boulding between negative and positive peace and followed by many others. See Kenneth Boulding, “Toward a Theory of Peace,” in Roger Fisher, ed., International Conflict and Behavioral Science (New York: Basic Books, 1964), pp: 70–87.

36 Monty Marshall and Keith Jaggers, “Polity IV Project,” Codebook and Data Files, 2000, www.bsos.umd.edu/cidcm/inscr/polity.

19 I N T R O D U C T I O N

two reasons. One is to tap into the ordinary association of “peace” with a condition of agreement and acceptance. The other is to identify peace with participation as the beginning of what can be a much more lasting and stable peace. The statistical association between peace and democ­racy may be U-shaped; both tough autocracies and well-established (usu­ally wealthy) democracies maintain civil peace. Semi-democracies (or anocracies) tend to be most prone to civil war.37 Participatory peace is thus a difficult status, one designed to measure whether the postwar state has entered a path toward democratic civil peace.

We are thus fully aware how challenging peacebuilding can be. Stable participatory polities usually reflect and rely upon a shared national iden­tity, well-functioning state institutions, a wide middle class, and a grow­ing economy. Both in part and often in whole, these are just what are missing in the typical post–civil war environment where there is often more than one ethnic identity, national identity is weak or contested, state institutions have been corrupted or destroyed altogether, the middle class is small (or has fled), and the economy has been geared to military production and the civilian economy (what there was in the first place) has been looted. Successful peacebuilding is the surprise, not the expec­tation.

Participatory peace is, however, worth striving for (hence measuring) because it offers the prospect of peace as a self-sustaining conflict resolu­tion mechanism—the promise that future disputes will be negotiated, re­solved according to constitutionally agreed procedures. Moreover, the likely alternatives seem worse. On the one hand, the destructiveness of civil anarchy is unacceptable both to all who suffer and to much of the international community forced to observe the suffering. Weak as the second sentiment is, it seems to be enough to launch peace operations when the destructiveness becomes overwhelming as it did in Somalia in 1992 and in Rwanda (after the genocide in 1994) or when the parties at last agree upon a peace. On the other hand, the humiliations and costs of international hierarchy make neocolonialism also unacceptable in the

37 Håvard Hegre, Tanja Ellingsen, Scott Gates, and Nils Petter Gleditsch, “Toward a Democratic Civil Peace?” American Political Science Review 95, no. 1 (March 2001): 33–48. Both democracy and wealth, which highly correlate in stable democracies, are do­ing the work, not democracy alone. See James D. Fearon and David D. Laitin, “Ethnicity, Insurgency and Civil War,” American Political Science Review 97, no. 1 (2003): 75–90. In related, ongoing research, Jennifer Gandhi and James R. Vreeland show that even in autoc­racies, political institutions can help reduce the risk of civil war. This would be consistent with our theoretical argument, if institutions are seen as ways to devolve some decision-making authority and increase participation, however marginally. See Jennifer Gandhi and James R. Vreeland, “Political Institutions and Civil War: Unpacking Anocracy,” Unpub­lished manuscript, Emory University and Yale University (August 30, 2004).

20 C H A P T E R 1

current age.38 Too many postcolonial peoples who have been subject to its indignities are determined to rule themselves. None of the potential imperial powers seem prepared to pay the military and economic costs of permanent rule. While we summarize and expand our earlier results on the determinants of peacebuilding, in this book we want a more finely textured discussion that reflects other aspects of success—the quality of public liberties, degrees of social integration, or the rates of economic growth, and these vary among our cases. So we present a systematic comparison based on statistical analysis and then a more nuanced dis­cussion of the quality of the peace in our case studies.39 (It does make a difference that a former guerrilla commander became the mayor of San Salvador.)

Distinguishing strategies and outcomes is another methodological challenge. In the real world, they are never completely separable: politi­cal strategies rely on expectations about expected outcomes. But the an­alytic separation should be highlighted as much as is feasible.40 For ex­ample, concerning the use of force—a UN-managed strategy of force is usually ineffective when it seeks to impose a peace (e.g., Somalia), but often effective when it is used in discrete bits to implement a comprehen­sive peace treaty (Cambodia, Eastern Slavonia). The difference here is the context of the comprehensive peace, not the same strategy with dif­ferent outcomes. Then the complications set in about how much coer­cion is compatible with “agreement” and so forth, and in the case stud­ies we will discuss this.

We also distinguish between fulfilling a mandate and establishing a peace. Peacebuilding, when comprehensively planned and executed, achieves a sustainable peace. But not all peace operations are well de­signed; some are stopgaps and others are misconceived. Peace operations can fulfill their specific mandates authorized by the Security Council or NATO or another body, and yet sustainable peace can still be elusive. Lightly armed peacekeepers sent into the middle of raging civil wars

38 For good discussions of the issue see Daniel Philpott, Revolutions in Sovereignty (Princeton: Princeton University Press, 2001); William Bain, Between Anarchy and Society: Trusteeship and the Obligations of Power (Oxford: Oxford University Press, 2003); and Nicholas Wheeler, Saving Strangers (Oxford: Oxford University Press, 2000).

39 Much of the statistical analysis is included in a supplement that we make available online (see chapter 3 for the web address), but we discuss the main results in chapter 3. This organizational structure of the book allows readers who are not interested in the tech­nical details to follow the argument without having to read through extensive technical discussions of data collection and coding, model estimation, and hypothesis testing.

40 If strategies are endogenous to expectations about outcomes, and we want to evalu­ate the effects of strategies on outcomes, this raises several technical issue with selection and endogeneity that we address in our statistical analysis in chapter 3 and in our supple­ment.

21 I N T R O D U C T I O N

where there is “no peace to keep” do their job, but in circumstances of frustration. Beleaguered peacekeepers and harried UN civil servants nat­urally want their performance to be measured by whether or not they fulfilled the mandate (to monitor a truce, deliver humanitarian supplies, hold an election, etc.) that they were given by the Security Council.41

This is reasonable and this we do in the case studies that follow. But we also want to assess in our statistical analysis whether the mandate itself is well designed to achieve a sustainable peace, and thus whether the Se­curity Council itself did its job.42 This we will measure also when we note how long the peace lasts after the peacekeepers leave. And, some­times, though rarely, peace operations can fail their mandates and fortu­nate domestic circumstances can rescue the peace, as happened when Angolan military forces killed opposition leader Jonas Savimbi and the intractable (to the UN) insurgency collapsed.

In our analysis of the UN’s role in making war and making peace, we will focus on four cases of failure—Somalia, Bosnia, Rwanda,43 and Cyprus—and six cases of success—Congo,44 El Salvador, Cambodia, Eastern Slavonia in Croatia, Brcko in Northern Bosnia, and East Timor. In the conclusion, we will highlight the lessons of these cases for under­standing the ongoing challenges of peace operations.

Some of these cases are well known, and we draw on available evi­dence to develop them in our analysis. Others we have observed in per­son and, in those cases, we draw on firsthand experience and primary research. Each case represents a particular type of failure or success, cho­sen to illustrate the key factors that our quantitative study and our theo­retical model reveal as important. Other cases will also be addressed in this book, though only in passing.

41 Many scholars follow them in this preference; see for example Simon Chesterman’s thoughtful and informative study of the variety of mandates: You the People (Oxford: Ox­ford University Press, 2004).

42 The Security Council has welcomed the Secretary-General’s Report, “No Exit without Strategy” (S/2001/394, April 20, 2001) in which sustainable peace is recognized as the goal toward which an exit strategy should be designed.

43 In our dataset, we code two events of civil war in Rwanda in the 1990s. The failure here refers to the peace process that started with the Arusha Accords in 1993 and ended with the genocide of 1994. The postgenocide peace process has been a mixed case: there have been genuine improvements in governance and reconciliation, but also significant vio­lence in border regions.

44 This is a good example of the difference between success conceived narrowly as the implementation of the mandate and a broader view of success that focuses on levels of vio­lence and participation in the country after the peacekeepers leave. With respect to both sovereign and participatory peace, the Congo was a peacebuilding failure according to cri­teria that we establish later in the book. But the UN mission in the Congo (ONUC) was successful in implementing its mandate of keeping the country together despite strong se­cessionist conflict in Katanga and elsewhere.

22 C H A P T E R 1

The book will focus on the 1990s because there was very little of this comprehensive peacebuilding before then.45 Our two Cold War cases— Congo and Cyprus—were exceptions. This was partly because the Cold War precluded UN involvement (due to Soviet or U.S. vetoes in the Secu­rity Council). Consequently, there emerged a backload of addressable civil wars, accounting for the surge in the early 1990s. Although the rate of onset of civil wars may be declining, the challenges of peacebuilding are all too likely to continue to arise.

One of the most important challenges the international community faces is thus the question of how to rebuild stable polities in the aftermath of civil war. How can the international community assist former combat­ants with a will to peace to prevent renewed hostility and to contain the ambitions of those who seek renewed civil war? What role should the in­ternational community play in ensuring that failed states do not relapse into chaos as soon as the international peacekeepers leave? The United Nations and various regional organizations, including NATO, have ac­cepted the responsibility to undertake “postconflict peacebuilding” and commissioned their member states to undertake extensive intrusions into the domestic affairs of other legally sovereign states.46

What guidelines should be developed to help steer these ambitious mandates? In current usage in the UN and among private voluntary or­ganizations, peacebuilding is an attempt, after a peace has been negoti­ated or imposed, to address the sources of present hostility and build local capacities for conflict resolution. Strengthening state institutions, increasing political participation, engaging in land reform, deepening civil society, finding ways to respect ethnic identities: all these are seen as ways to improve the prospects for peaceful governance.47 In pluralistic

45 However, in our statistical analysis, we use data from the entire post-1945 period to evaluate the UN’s record. We tested for significant differences between the pre–Cold War and post–Cold War periods, but the number of cases becomes too small when we break the postwar period in this way, so we prefer to analyze all years since 1945.

46 For a discussion of the concept, see Goulding 1993: 451–64; Thomas Franck, “A Holistic Approach to Peace-building,” in Olara Otunnu and Michael Doyle, eds., Peace­making and Peacekeeping for the New Century (Lanham, MD: Rowman and Littlefield, 1998); and Elizabeth Cousens, Chetan Kumar, and Karin Wermester, eds., Peacebuilding as Politics: Cultivating Peace in Fragile Societies (Boulder, CO: Lynne Rienner 2000). The UN’s own views can be found in Boutros-Ghali 1992); and Kofi Annan, The Causes of Conflict and the Promotion of Durable and Sustainable Peace in Africa: Report of the Secretary-General (New York: United Nations, 1998).

47 For a discussion of comprehensive peace, see Boulding 1964; and Arie Kacowicz, Peaceful Territorial Change (Columbia: University of South Carolina Press, 1994), chap. 1. For a valuable collection of papers on peacebuilding see Cousens, Kumar, and Wermester 2000; UN Department for Development Support and Management Services and UN Indus­trial Development Organization (1995); and Gareth Evans, Cooperating for Peace (Lon­don: Allen and Unwin, 1993).

23 I N T R O D U C T I O N

societies, conflicts are inevitable. The aim of peacebuilding is to build the social, economic, and political institutions and attitudes that will prevent the inevitable conflicts that every society generates from turning into vio­lent conflicts.48 In effect, peacebuilding is the front line of preventive ac­tion.

Plan of the Book

We begin in chapter 2 with the development of a theoretical framework that explains how peacekeeping can help achieve sovereign and partici­patory peace after civil war. We start with an analytical review of the de­bate on the causes and cures of civil wars—the primary strategic element in which the UN found itself in the 1990s. Then we develop the logic be­hind our concept of the peacebuilding triangle that explains how the in­ternational community could become involved effectively. The chapter highlights the ways in which international peacekeeping and peace en­forcement assistance can compensate for two key barriers to building peace: on the one hand, the hostility that civil wars generate and, on the other hand, the lack of local capacity that makes political and economic reconstruction so difficult. The argument presented here thus identifies the centrality of an international role in resolving civil war conflicts and the key elements essential for successful peacebuilding.

In chapter 3, we draw on a data set we have constructed that includes all civil wars since 1945 in order to analyze the determinants of success­ful peacebuilding after civil war. Here, we assess the effectiveness of UN peace operations by comparing peacebuilding outcomes in cases with and without a UN intervention. This macrolevel analysis demonstrates the centrality of the peacebuilding triangle and identifies the kinds of roles that the international community has played when peacebuilding has been successful.

Drawing on examples from Somalia and Bosnia, chapter 4 offers a mi­crolevel case study analysis of how and why the UN tends to fail at mak­ing war—imposing by force an overall settlement of civil conflict. Here we also discuss the exceptional case of the Congo in 1960–65, where for special reasons the UN succeeded in imposing an (albeit in many ways inadequate) peace. Chapter 5 explores, also at the microlevel, how the UN has succeeded in making peace in countries as various as El Sal­

48 No peace is perfect. Isaiah prophesied that we shall know peace when we see the lamb lie down with the lion. The American comedian Woody Allen has added a valuable warn­ing for our world: one of the two might not get much sleep. Isaiah 11:6 and Woody Allen, Without Feathers (New York: Warner Books, 1976) p. 28.

24 C H A P T E R 1

vador, Cambodia, Croatia Bosnia (post-Dayton), and East Timor. And chapter 6, focusing on the long drawn out peacekeeping effort in Cyprus and the horrendous tragedy in Rwanda, explains how peacekeeping has sometimes failed. Five and six focus on strategies and management and the key role transitional authority plays in managing the trade-offs of the peacebuilding triangle.

The case studies expand the scope of the statistical analysis by allow­ing us to explore the distinction between macrolevel and microlevel peacebuilding success that we do not explore in the statistical analysis. In the cases, we focus on microlevel success and on variables that are hard to measure and use in a large-N statistical analysis; while in the statistical analysis, we focus on macrolevel success and analyze the effects of UN operations while controlling for other factors. In the macrolevel statistical analysis we are able to assess the effectiveness of the UN be­cause we can compare civil wars with UN involvement to those civil wars where the UN did not intervene. The case studies complement that analysis by focusing more closely on cases of UN intervention and giving us a better view of the dynamics between the factions and UN missions.

While the statistical analysis includes all peace processes since 1945 and helps us discuss correlations between peacebuilding outcomes and types of UN missions, the case studies help trace the mechanisms through which UN peace operations contribute to successful peacebuilding. Here we focus on the processes of institutional transformation and capacity ex­pansion that increase the costs of noncooperation for the factions and create incentives for them to keep the peace. We focus on the peacekeep­ers’ implementation of their mandate, identifying particular successes and failures in each case, and offering a perspective of the process of peace-building over time, discussing both the period before and after the UN’s involvement.

The case studies also help us identify possible explanations for partic­ular peacebuilding outcomes that are not always captured by our model. These can be idiosyncratic explanations—which do not concern our tri­angle model of peacebuilding—or they can be more important, general­izable explanations that our model overlooks. To identify explanations that go beyond our model, but also to place our discussion of the UN’s missions in context, we begin each case with a brief history of the con­flict and introduce the main actors as well as the circumstances that led to the UN’s involvement. We present “vital statistics” for each case for all variables that we consider in our statistical analysis so that readers can get an immediate sense of how this case fits with the average case in our dataset. We also discuss explicitly how well (or how poorly) those cases fit the statistical model and explain why. We consider alternative explanations and link our analysis of the success or failure of peace­

25 I N T R O D U C T I O N

building to the theoretical explanations of civil war onset and recurrence developed in the theory chapter.

In chapter 7, we compare how various strategies to make, keep, build, and enforce peace were implemented and illustrate the role transitional authority plays in managing a transition to peace. Here, we develop the concept of ecologies of peacebuilding and return to our case studies and our data to characterize the particular peacebuilding ecology for each of our cases. This allows us to better evaluate the type of UN involvement that we should have observed in each case if the UN mandated and re-sourced its missions efficiently. Finally, in the conclusion, we consider al­ternatives to UN peacebuilding and summarize the record of experience from which the organization is still learning today.

If the world does not fall back into another cold war among the per­manent members of the Security Council, then the UN is likely to be in­volved in the new civil and international wars. Anthony Lake, the Clin­ton administration national security adviser, expressed it well in his 6 Nightmares: “America must also do its part in peacekeeping operations, working whenever possible through the United Nations. If we do not, our interests suffer, our leadership diminishes, and innocent people die.”49 After much initial skepticism about the value of peacebuilding, the Bush administration (following 9/11) launched two extremely ambi­tious efforts to remake Afghanistan and Iraq, both eventually with UN involvement.

The CIA’s Global Trends 2015 study presented a comprehensive sum­mary of the prospective threat and is worth quoting at length:

Through 2015, internal conflicts will pose the most frequent threat to stability around the world. . . . Many internal conflicts, particularly those arising from communal disputes, will continue to be vicious, long-lasting and difficult to terminate—leaving bitter legacies in their wake. They frequently will spawn internal displacements, refugee flows, humanitarian emergencies, and other regionally destabilizing dislocations. If left to fester, internal conflicts will trig­ger spillover into inter-state conflicts as neighboring states move to exploit op­portunities for gain or to limit the possibilities of damage to their national interests. . . .

The United Nations and several regional organizations will continue to be called upon to manage some internal conflicts because major states—stressed by domestic concerns, perceived risk of failure, lack of political will, or tight resources—will wish to minimize their direct involvement.50

49 Anthony Lake, 6 Nightmares: The Real Threats to American Security (Boston: Little, Brown, 2000), p. 284.

50 http://www.odci.gov/cia/publications/globaltrends2015/index.html#link3, January 5, 2001.

26 C H A P T E R 1

Despite the stresses of the global war on terrorism, the Global Trends 2015 around the world today. Indeed, a new set of peacebuilding challenges— with or without the UN—are on the horizon, reflecting if nothing else the ambitious agenda of disarmament-through-regime-change embodied in the new U.S. National Security Doctrine of preventive “preemptive defense.” Unfortunately, it thus appears, the lessons that this book will be drawing will likely be useful both for the present and the future.